Keywords: abdominal–inguinal skin, cremaster muscle, lumbar spinal nerves, motoneurones, pyramidal muscle
 |  |
| ||||
Copyright © Anatomical Society of Great Britain and Ireland 2002 An anatomical and electrophysiological study of the genitofemoral nerve and some of its targets in the male rat Correspondence Dr Rosa Angélica Lucio, Centro de Investigaciones Fisiológicas, Universidad Autónoma de Tlaxcala, Carretera Tlaxcala-Puebla Km 1.5, Código Postal 90070, Tlaxcala, México, Apartado Postal 262. Tel.: (52 246) 4621557; fax: (52 246) 4621557; e-mail: ralucio/at/cci.uatx.mx. Accepted September 24, 2002.  This article has been cited by other articles in PMC. | |||||||||||
Abstract Anatomical descriptions of the genitofemoral nerve (GFn) innervating the lower pelvic area are contradictory. Here we re-examine its origin and innervation by its various branches of principal target organs in the male rat. Using gross dissection, electrophysiological techniques and retrograde tracing of motoneurones with horseradish peroxidase, we confirm that the GFn originates from lumbar spinal nerves 1 and 2, and that at the level of the common iliac artery it divides into a lateral femoral and a medial genital branch. In contrast to previous studies, we report that the genital and not the femoral branch innervates the abdominal–inguinal skin, and not only the genital but also the femoral branch innervates the cremaster muscle (Cm) surrounding the testes. Motoneurones innervating the Cm proper are located in the ventral nucleus of L1 and L2, and those innervating the muscular transition region of the rostral Cm are located in the ventral nucleus in L1 and the ventrolateral nucleus in L2. The GFn may contribute to male reproductive performance by transmitting cutaneous information during copulation and, via contraction of the Cm to promote ejaculation, the protective displacement of the testes into the abdominal cavity during fighting and as a sperm-protecting thermoregulatory measure. Keywords: abdominal–inguinal skin, cremaster muscle, lumbar spinal nerves, motoneurones, pyramidal muscle | |||||||||||
Introduction The pelvic area of mammals is a region of great anatomical and functional complexity. However, despite its vital role in the performance of eliminative and reproductive functions, our understanding of its organization is far from complete, and where information does exist it is often unclear and contradictory. This is illustrated by continuing efforts – largely in the male rat – to describe the composition, distribution and function of the peripheral pelvic nerves. While consensus exists on major nerves such as the pelvic (Langworthy, 1965; Dail et al. 1975; Hulsebosch & Cogeshall, 1982; Lucio et al. 1994; Hubscher & Johnson, 2000), the pudendal (Larsson & Södersten, 1973; Hulsebosch & Cogeshall, 1982; McKenna & Nadelhaft, 1986, 1989; Pacheco et al. 1997), and the hypogastric (Langworthy, 1965; Hulsebosch & Cogeshall, 1982; Dail et al. 1985, 1989; Claire & Bradley, 1998; Hubscher & Johnson, 1999, 2000), reports on the genitofemoral nerve (GFn) remain contradictory. In the male rat the GFn has been variously reported to arise from lumbar spinal nerves 2 and 3 (Greene, 1955; Clegg & Doyle, 1966; Grant, 1966), and lumbar spinal nerves 1 and 2 (Hodson, 1970; Hebel & Stromberg, 1986), confirmed more recently by the use of immunohistochemical techniques (Nagy & Senba, 1985; Wang et al. 1989; Kar et al. 1990). Running caudally, the lumbar spinal nerves converge to form the GFn proper. In all species studied to date including the rat (Greene, 1955; Hebel & Stromberg, 1986), dog (Getty & Hadek, 1964), horse (Sisson & Grossman, 1968) and human (Wilson & Wilson, 1978; Hollinshead & Rosse, 1985), the GFn runs caudally parallel to the descending aorta, passes through the psoas major muscle, and divides at the level of the common iliac artery to form genital and femoral branches (Greene, 1955; Hodson, 1970). The femoral branch has been reported to innervate the inguinal skin in the rat (Greene, 1955; Clegg & Doyle, 1966; Hebel & Stromberg, 1986) and humans (Hollinshead & Rosse, 1985; Reid & Cros, 1999) and the scrotal skin in the rat (Clegg & Doyle, 1966), and the genital branch also to innervate the scrotal skin in the rat (Greene, 1955; Clegg & Doyle, 1966; Nagy & Senba, 1985; Hebel & Stromberg, 1986) and humans (Reid & Cros, 1999), the cremaster muscle (Cm) surrounding the testes in the rat (Greene, 1955; Clegg & Doyle, 1966; Grant, 1966; Adams et al. 1967; Hebel & Stromberg, 1986) and humans (Hollinshead & Rosse, 1985; Reid & Cros, 1999), the spermatic cord and vaginal tunica of the testes in the rat (Durwood, 1964; Miller et al. 1964), and the dartos muscle lining the scrotal sac in humans (Mitchel, 1938; Clegg & Doyle, 1966). However, more recently it has been reported in the rat that the scrotum is innervated by branches of the pudendal nerve (Pacheco et al. 1997; McKenna & Nadelhaft, 1986) and not by the GFn. Adding to this complexity are reports that the GFn is composed of both motor and cutaneous components in the cat and rabbit (Langley & Anderson, 1896), the rat (Greene, 1955; Hodson, 1970) and humans (Reid & Cros, 1999). Given continued uncertainty as to the course and targets of the GFn it was the aim of the present study to re-examine its origin and the distribution of its branches in the male rat, paying particular attention to the pattern of innervation of the Cm and abdominal–inguinal skin. We report that the genital and not the femoral branch innervates the abdominal–inguinal skin, and not only the genital but also the femoral branch innervates the Cm. We also report that contraction of the Cm displaces the testes, and that the Cm is innervated by motoneurones located in the ventral nucleus of the grey matter of spinal segments L1 and L2. | |||||||||||
Materials and methods Adult male Wistar-strain rats weighing 300–400 g were used. They were kept in groups of five in acrylic cages (47 × 33 height, 20 cm), on a 12 : 12 h LD cycle (lights on at 22 : 00 h), at 24 ± 2 °C and with pelleted rodent food (Purina, Mexico) and water available continuously. In Experiment I and II animals were anaesthetized with an intraperitoneal injection of urethane (1.6 g kg−1, 20% in distilled water), and killed with an overdose of urethane at the end of the experiment. In Experiment III animals were anaesthetized with an intraperitoneal injection of sodium pentobarbital (26 mg kg−1 body weight; Anestesal, Smith Kline, Mexico), and killed during perfusion at the end of the experiment. | |||||||||||
Experiment I: anatomical description Using gross dissection, it was the purpose of this part of the study to identify the origin and trajectory of the GFn, including innervation by its various branches of principal target organs. Materials and methods Twenty rats were used. They were anaesthetized and placed on their back, their ventrum shaved, a 8–10-cm midline incision made in the abdominal wall, the viscera externalized, and tissue kept moist with gauze pads soaked in warm 0.09% saline solution. Using the anatomical description of Greene (1955) and a Nikon stereoscopic microscope (SMT-2T) the descending aorta was located, and parallel to it the readily identifiable GFn. At the level of the renal artery (see Fig. 2) the GFn was followed rostrally until the vertebral column was reached in order to determine the origin of the nerve. From the renal artery the GFn was then followed caudally in order to determine its target structures.
Results and discussion Essentially the same pattern of results was obtained in all 20 animals. The GFn appeared to be formed from fibres originating from the first lumbar spinal nerve which emerges between the first and second lumbar vertebrae, and the second spinal nerve which emerges between the second and third lumbar vertebrae (Fig. 1). After leaving the vertebral column the two spinal nerves typically formed two anastomotic branches, although in two animals only one was seen. Following this, a distinct GFn could be identified Fig. 1 which ran caudally and pierced the psoas major muscle not far from the midline.
The GFn continued caudally close to the descending aorta and dorsal to the renal artery, followed the common iliac artery for a short distance and then formed two branches (Fig. 2). Both of these, the lateral or femoral branch and the medial or genital branch, continued caudally, each giving rise to three main ramifications (Figs 2 and 3).
Femoral branch Following the femoral branch along the external iliac artery (Fig. 2) a first laterally running branch was observed in the rostroventral region characterized by the confluence of fibres from the oblique internal muscle (OIm), the transverse abdominal muscle (TAm) and initial fibres of the Cm (Figs 3 and 4). We designated this the muscular transition region (mtr). No tendons could be seen and it was only possible to distinguish the muscles by a change in the direction of the fibres. Thus, the superficially lying fibres of the OIm ran mediolaterally, with progressively more caudal fibres running circularly to form the external layer of the cremasteric sac (Fig. 4). Fibres of the TAm initially ran caudally and mediolaterally and then changed direction slightly to distribute longitudinally and meet with the first fibres forming the internal layer of the cremasteric sac (Fig. 4). A change in the direction of the two sets of fibres and in the width of the two muscles was considered to represent the start of the Cm.
The second ramification of the femoral branch turned to innervate the rostral Cm, with fine nerve fibres distributed across the dorsal and ventral surfaces of the muscles. The third ramification ran caudolaterally, distributing fibres across the lateral face of the Cm (Fig. 3). Genital branch Following the genital branch in a caudal direction close to the internal iliac artery it was observed to give off a first medial ramification which crossed the Cm in a mediocaudal direction and then turned ventrolaterally to innervate the abdominal–inguinal skin (Figs 2 and 3). Before turning ventrolaterally it gave off a small medial branch which appeared to innervate the pyramidal muscle (Pm) located ventrally to the beginning of the abdominal muscles. A second ramification appeared to innervate the dorsal and medial regions of the Cm, and the third ramification appeared to innervate the ventral region.In summary, while the present findings agree with previous reports that the GFn richly innervates the Cm, they indicate that the femoral as well as the genital branch innervates this muscle. The results also suggest that the genital branch transmits sensory information from the inguinal skin, and not the femoral branch as previously reported (Greene, 1955; Clegg & Doyle, 1966; Hebel & Stromberg, 1986; Reid & Cros, 1999). | |||||||||||
Experiment II: electrophysiological corroboration Using dissection to describe the innervation by nerves of their target organs it is usually not possible to identify their finest terminations or to distinguish these from commonly occurring passing fibres. It was therefore our purpose in this second experiment to verify the findings of Experiment I using electrophysiological techniques. Materials and methods Thirty rats divided into three groups were used. They were anaesthetized, and a large midline abdominal incision made to visualize the GFn.Recording afferent electrical activity Group A (n = 10) Electrical activity of the GFn and its femoral and genital branches was recorded using bipolar silver chloride hook electrodes connected to a Grass 7P511 pre-amplifier, and the activity displayed on a Tektronix 2214 storage oscilloscope connected to a PC computer containing wave form software (Grabber II). To identify the sensory field of the GFn on the skin, electrical activity was recorded in the nerve just before it divided into the femoral and genital branches, while brushing a cotton bud across the scrotum and the femoral, perineal, abdominal and inguinal areas. Once a response was registered the sensory field was mapped by applying stimulation to the skin using Von Frey fibres of 1.20–1.80 mg force in ascending magnitude and at intervals of 0.5 cm. This was done with the GFn intact and after sectioning it above the electrode to avoid recording possible efferent activity. In the same preparation, activity in response to brushing across the sensory field was then registered in the femoral branch and in the genital branch just caudal to the bifurcation of the GFn. To verify the lack of activity observed in the femoral branch (see below), this was sectioned and activity again recorded in the GFn rostral to its bifurcation. As this confirmed the activity registered in the GFn to be transmitted by the genital branch, activity was again recorded in this branch after sectioning each of its subbranches. Recording efferent electrical activity Group B (n = 10) To confirm the identity of the striated muscles described in Experiment I as being innervated by the GFn, electromyograms (EMG) were recorded from the mtr, Pm and Cm during electrical stimulation of the GFn or its femoral and genital branches. Stimulation was applied using bipolar silver chloride electrodes (2 mm diameter) connected to a Grass SIU5 stimulus isolation unit activated by a Grass S48 stimulator. Square pulses of 0.1–0.2 ms duration and varying in intensity from threshold (T) to 2 × T were applied. EMGs were recorded using 0.1-mm-diameter stainless-steel bipolar electrodes inserted into the muscle and connected to a Grass 7P511 preamplifier. EMGs were displayed on a Tektronix 2214 storage oscilloscope connected to a PC computer containing waveform transfer software (Grabber II). Contraction of the muscles was recorded while applying stimulation to the intact GFn before its bifurcation, and after sectioning it before its bifurcation and stimulating the femoral and then the genital branches. Contraction of the Cm Group C (n = 10) To investigate more closely the role of the GFn in regulating contraction of the Cm, the GFn was visualized as described above and a longitudinal incision made in the scrotum to expose the Cm. Stimulation was applied to the GFn rostral to its bifurcation and the contractile force generated in the Cm measured using a Grass FT03C tension transducer attached to the most caudal extremity of the Cm. Tension was measured during application of square pulses of 0.1 ms and 2 × T at frequencies varying between 50 and 120 Hz. Results and discussion Group A Essentially the same pattern of response was recorded in all 10 animals. No activity was registered in the GFn in response to brushing the scrotal, femoral or perineal skin whereas clear activity was seen when brushing the caudal abdominal and medial inguinal skin. The response was maintained throughout stimulation, indicating tonic activation. Applying finer stimulation with the Von Frey fibres showed this area of sensitivity to extend approximately 2.5 cm rostrally and 2.0 cm laterally from the perineal area (Fig. 5a). The response threshold was registered at 1.37 mg and the maximal response at 1.73 mg. Equivalent stimulation of the skin contralateral to the side of recording failed to elicit any response in the GFn.
Whereas no activity in response to stimulation was recorded in the femoral branch (Fig. 5b1), clear activity was recorded in the genital branch (Fig. 5b2, b3), including after the femoral branch had been cut. Group B Electrical stimulation (2 × T) of the GFn resulted in contraction of the mtr, Pm and Cm.mtr Stimulation of the femoral branch elicited strong contractions in this region. The EMG (Fig. 6a) indicated a complex response with a latency of 1.5 ms, amplitude of 3 V and a duration of approximately 4 ms. Stimulation of the genital branch (Fig. 6b) resulted in a weaker response with a latency of 2 ms, amplitude of 1.5 V and a duration of approximately 2 ms.
Pm Whereas stimulation of the femoral branch failed to elicit contraction of the Pm (Fig. 7a), stimulation of the genital branch provoked a complex response (Fig. 7b).
Cm The strongest contractions in response to stimulation of the GFn were seen in the Cm. Since this muscle is large we recorded activity from rostral, medial and caudal regions. Stimulation of the femoral branch resulted in contraction in all three regions, with similar latencies of 1.5 ms, amplitude of 10 V and duration of 2 ms (Fig. 8a). Stimulation of the genital branch resulted in somewhat stronger contractions of the Cm, with latencies of 1.5 ms, amplitude of 15 V and duration of 2 ms for all three regions(Fig. 8b). As can be seen from the EMG recordings in Fig. 8, stimulation of the genital branch resulted in a more complex response than stimulation of the femoral branch.
Group C Electrical stimulation of the GFn (2 × T) resulted in contraction of the Cm and in movement of the ipsilateral testis but not in its displacement. The same stimulation but at 50 Hz resulted in displacement of the testis to the rostral region of the Cm, while the same form of stimulation at 120 Hz resulted in displacement of the testis into the abdominal cavity (Fig. 9a). When stimulation stopped, the testis descended to its original position in the caudal cremasteric sac. The force generated by the Cm during stimulation at 50 Hz was 60 mN accompanied by incomplete tetanization, and during stimulation at 120 Hz it was 120 mN, reaching complete tetanization (Fig. 9b).
In summary, the findings corroborate the anatomical description of Experiment I. Thus, we again found that the genital branch transmits sensory information from the area of abdominal–inguinal skin innervated by the GFn and not the femoral branch as previously reported for the rat (Greene, 1955; Hebel & Stromberg, 1986) and humans (Hollinshead & Rosse, 1985; Reid & Cros, 1999), and that the GFn does not innervate the scrotal skin by any of its branches as was previously thought (Greene, 1955; Clegg & Doyle, 1966; Nagy & Senba, 1985; Hebel & Stromberg, 1986; Reid & Cros, 1999). In addition, we report for the first time that the femoral branch principally innervates the mtr, and that a subbranch of the genital branch innervates the Pm while further genital branches profusely innervate the Cm across its dorsal and ventral face. | |||||||||||
Experiment III: spinal location of motoneurones Given that the GFn originates from two spinal segments (L1 and L2) and that it innervates the Cm via its femoral and genital branches and the mtr via its femoral branch, we were interested to investigate the location in the spinal cord of the motoneurones innervating these two muscles. Using retrograde transport of horseradish peroxidase coupled to wheat germ agglutinin (HRP-WGA) we asked whether the spinal distribution and morphometric characteristics of the motoneurones innervating the two muscles differed in any way. Materials and methods Animals and HRP-WGA retrograde labelling Eighteen rats were divided into two groups, Cm males (n = 9) and mtr males (n = 9). They were anaesthetized and a scrotal longitudinal incision was made to visualize the Cm, or an abdominal midline incision to visualize the mtr. Under a Nikon (SMT-2T) surgical microscope the left Cm or the left mtr was injected with a total of 5µL of 5% HRP-WGA applied using a 10-µL Hamilton syringe. In the case ofthe Cm, each rat received three injections at different sites and in the case of the mtr only two. To minimize diffusion of the tracer, the needle was left in situ for 1–2 min after each approximately 15 s injection.After 48 h survival time, animals were perfused intracardially with 500 mL physiological saline followed by 800 mL fixative at room temperature (1% paraformaldehyde, 0.5% glutaraldehyde, 0.002 CaCl2, and 0.1 m sucrose in 0.1 buffer phosphate, pH 7.3). Forcryoprotection 500 mL of 10% sucrose in 0.1 m buffer phosphate, pH;7.3 at 4 °Cwas used, and thoracic (T) and lumbar (L) segments of the spinal cord (T13–L3) then removed and kept in the sucrose solution for at least 24 h before sectioning. After embedding in Tissue-Tek®medium for frozen specimens, the spinal cord segments were cut (50 µm) longitudinally (Cm males, n= 2; mtr males, n= 2) and transversally (Cm males, n= 7; mtr males, n= 7) using a cryostat. Floating sections were rinsed in 0.0025 M acetate buffer (pH 3.3) for 20 min, pre-incubated in a solution containing 341.2 mg sodium nitroferricyanide 2-hydrate, 15 mg tetramethylbenzidine, 7.5 absolute ethanol in 292.5 mL 0.00025 m acetate buffer (pH 3.3), and incubated in the dark for 20 min in 300 mL pre-incubation solution containing 0.2 mL 30% H2O2 (Mesulam, 1978). Sections were then rinsed in 0.0025 M acetate buffer (pH 3.3), mounted on gelatinized slides, air dried, briefly dehydrated in graded ethanols (70°, 96° and absolute; 10 s in each), cleared in xylene and mounted in Cytoseal 60™. The enzymatic reaction was performed in a dark room under faint light to improve HRP-WGA visibility. Measurements and image processing Longitudinal sections were necessary to determine which segments of the spinal cord presented labelled motoneurones and to determine the extent of the motoneurone pool. In addition, the number and diameter of labelled somas was recorded so as to correct cell counts made in transverse sections using the correction factor of Abercrombie (1946). Two additional males were used to stain spinal cord sections using the Nissl method so as to identify the location of the motoneurone pool observed in the HRP-WGA-labelled sections.Sections were evaluated by an observer using coded slides so as to be blind to the muscle which had been injected, and they were photographed using an Olympus BH2 microscope. For the analysis of HRP-WGA-labelled sections we used a computer connected to an Optiphot-2 Nikon microscope equipped with a colour video camera (TK-C1380; JVC) and a Sony frame memory unit. We used the Neurograph computer system (Microptic, Spain) to plot the distribution of labelled neurones in outlines of longitudinal sections and to perform the morphometric analysis. For this, measurements were only made for cells whose soma and dendritic length could be clearly distinguished from neighbouring cells. Soma area was measured by tracing the outline of the labelled cell body and calculating the area enclosed by this, and the area of primary dendritic arborization was measured by anchoring points with the computer mouse at the tip of each primary labelled dendrite, connecting these with a line and calculating the area enclosed by this. Primary dendrite length was measured by tracing a single line with the mouse along the dendrite image. Results and discussion As shown in Table 1, a substantial number of labelled cells was observed in all animals following injection of HRP-WGA in either muscle.
Cm motoneurones In all cases, labelled neurones were found in the ventral nucleus of lamina IX (transverse sections), in the mediocaudal region of L1 and in the mediorostral region of L2 (longitudinal sections; Fig. 10) ipsilateral to the side of injection. The neurone pool extended 2026.99 ± 21.75 µm, with a somewhat higher percentage of neurones located in L2 (Table 1).
The transverse Nissl-stained sections confirmed the labelled motoneurones to be confined to the ventral nucleus. Both in the transversal and in the longitudinal sections the cell somas had a conical or rounded appearance, and in the longitudinal sections were closely and evenly spaced, with their dendrites intermixed. The dendrites were orientated dorsolaterally and dorsomedially to the grey matter (Fig. 10). Table 1 gives the raw count of the mean number of labelled neurones per animal and after applying the correction factor of Abercrombie (Table 1). mtr motoneurones In all cases, labelled neurones were found in the ventral nucleus of lamina IX (transverse sections), throughout the length of L1 (longitudinal sections), and in the ventrolateral nucleus of lamina IX (transverse sections), in the rostromedial region of L2 (longitudinal sections; Fig. 11) ipsilateral to the side of injection. The neurone pool extended 2158.66 ± 123.61 µm, with the highest percentage of neurones located in L2 (Table 1).
The transverse Nissl-stained sections confirmed the labelled motoneurones to be located in two distinct nuclei, one ventral in L1 and the other ventrolateral in L2. Both in the transversal and in the longitudinal sections the cells had a stellate appearance, and although in the longitudinal sections they were well spaced their extensive dendritic arbors overlapped. These long dendrites were orientated particularly to the dorsolateral region of the grey matter (Fig. 11). Table 1 gives the raw count of the mean number of labelled neurones per animal and after applying the correction factor of Abercrombie. Returning to the aim of this experiment, the findings confirm that the Cm and mtr are innervated by motoneurones located in spinal segments L1 and L2, with a somewhat higher proportion of neurones in L2. The results also reveal several differences between the populations of neurones innervating each of the muscles. Whereas those innervating the Cm were found only in the ventral nucleus, mtr motoneurones were located both in the ventral and in the ventrolateral nuclei. The number of neurones innervating the Cm was somewhat less than for the mtr (mean ± SE; 57.00 ± 1.90 vs. 74.90 ± 1.30), the area of their somas was somewhat bigger (483.60 ± 32.10 vs. 409.10 ± 12.5) and the area of primary dendritic arborization was less than for the mtr motoneurones (4867.30 ± 1433.10 vs. 11700.10 ± 1029.50). Although the dendrites of the motoneurones innervating the Cm were relatively short, due to the higher density of these cells there was considerable dendritic overlap. An unexpected finding was the conspicuous tendency observed in the long dendrites of mtr motoneurones to orientate towards the dorsolateral region of the grey matter. To our knowledge this is the first time such findings have been reported, and they would seem to imply differences in the manner of functioning of the Cm and mtr. | |||||||||||
Discussion The results of the present study show the GFn to be a nerve of considerable complexity and suggest its involvement in several aspects of male reproductive function. While agreeing with previous reports that the GFn originates from spinal nerves L1 and L2, and that at the level of the common iliac artery it bifurcates to form femoral and genital branches, the present findings differ from previous reports in several respects. In contrast to previous studies (see Introduction) we found that the genital and not the femoral branch innervates the abdominal–inguinal skin, and not only the genital but also the femoral branch richly innervates the Cm, the principal target structure of the GFn. In addition, as shown by the results of Experiment III, retrograde tracing of the motoneurones innervating the Cm proper and associated mtr revealed a differentiated pattern of the neurone location and morphology in the spinal cord, suggesting functional specificity. Differences between the present findings and previous studies are not surprising. Most previous reports on the GFn are part of larger anatomical descriptions of the entire animal (Greene, 1955; Grant, 1966; Hebel & Stromberg, 1986) or of the general pelvic area (Langworthy, 1965; Clegg & Doyle, 1966), and so do not give detailed attention to the finer distribution of the GFn. Since this was our sole purpose and the findings we obtained from dissection, electrophysiological recordings and retrograde tracing of motoneurones were consistent, we consider that the present results present a reliable neuroanatomical picture. More speculatively, we may briefly consider what these findings might mean in functional terms. Apart from the possibility that the genital branch transmits somatosensory input from the abdominal–inguinal skin stimulating and helping the male orientate during copulation, the rich and differentiated innervation of the cremasteric region seems particularly important. Thus, differences in the origin and morphology of motoneurones innervating the Cm and mtr might serve different functions: in the case of the Cm, strong synchronized contraction facilitating ejaculation (Lucio et al. 2001), protective displacement of the testes into the abdominal cavity during fighting or for thermoregulatory purposes, and in the case of the mtr, a sphincter-like function maintaining the testes within or outside the abdominal cavity. Based on the present anatomical description it is now possible to test the effect of transecting the various branches of the GFn on male reproductive performance in a more precise manner (Lucio et al. 2001). | |||||||||||
Acknowledgments We thank Jorge Rodríguez and Carolina Rojas for excellent technical assistance. This work was supported by a grant (CONACyT 2283 PN to RAL). | |||||||||||
References
| |||||||||||
